Few-shot Named Entity Recognition with Cloze Questions

TL;DR

This paper introduces a novel few-shot NER method using cloze-style questions and pattern-based rephrasing, significantly reducing the need for annotated data and outperforming standard fine-tuning on biomedical datasets.

Contribution

It adapts Pattern-Exploiting Training with cloze questions for NER, enabling effective few-shot learning without manual annotations or distant supervision.

Findings

Outperforms standard fine-tuning in few-shot NER tasks

Achieves comparable or better results than other few-shot baselines

Effective on biomedical datasets with limited labeled data

Abstract

Despite the huge and continuous advances in computational linguistics, the lack of annotated data for Named Entity Recognition (NER) is still a challenging issue, especially in low-resource languages and when domain knowledge is required for high-quality annotations. Recent findings in NLP show the effectiveness of cloze-style questions in enabling language models to leverage the knowledge they acquired during the pre-training phase. In our work, we propose a simple and intuitive adaptation of Pattern-Exploiting Training (PET), a recent approach which combines the cloze-questions mechanism and fine-tuning for few-shot learning: the key idea is to rephrase the NER task with patterns. Our approach achieves considerably better performance than standard fine-tuning and comparable or improved results with respect to other few-shot baselines without relying on manually annotated data or distant supervision on three benchmark datasets: NCBI-disease, BC2GM and a private Italian biomedical corpus.